It is well known the external surface of airfoils may be cooled by conducting cooling air from an internal cavity to the external surface via a plurality of small passages. It is desired that the air exiting the passages remain entrained in the boundary layer on the surface of the airfoil for as long a distance as possible downstream of the passage to provide a protective film of cool air between the hot mainstream gas and the airfoil surface. The angle which the axis of the passage makes with the airfoil surface and its relation to the direction of hot gas flow over the airfoil surface at the passage breakout are important factors which influence film cooling effectiveness. Film cooling effectiveness E is defined as the difference between the temperature of the main gas stream (T.sub.g) and the temperature of the coolant film (T.sub.f) at a distance x downstream of the passage outlet, divided by the temperature difference between the temperature of the main gas stream and the coolant temperature (T.sub.c) at the passage outlet (i.e., at x=0) thus, E=(T.sub.g -T.sub.f)/(T.sub.g -T.sub.c). Film cooling effectiveness decreases rapidly with distance x from the passage outlet. Maintaining high film cooling effectiveness for as long a distance as possible over as large a surface area as possible is the main goal of airfoil film cooling.
It is well known in the art, that the engine airfoils must be cooled using a minimum amount of cooling air, since the cooling air is working fluid which has been extracted from the compressor and its loss from the gas flow path rapidly reduces engine efficiency. Airfoil designers are faced with the problem of cooling all the engine airfoils using a specified, maximum cooling fluid flow rate. The amount of fluid which flows through each individual cooling passage from an internal cavity into the gas path is controlled by the minimum cross-sectional area (metering area) of the cooling passage. The metering area is typically located where the passage intersects the internal cavity. The total of the metering areas for all the cooling passages and orifices leading from the airfoil controls the total flow rate of coolant from the airfoil, assuming internal and external pressures are fixed or at least beyond the designer's control. The designer as the job of specifying the passage size and the spacing between passages, as well as the shape and orientation of the passages, such that all areas of the airfoil are maintained below critical design temperature limits determined by the airfoil material capability, maximum stress, and life requirement considerations.
Ideally, it is desired to bathe 100% of the airfoil surface with a film of cooling air; however, the air leaving the passage exit generally forms a cooling film stripe no wider than or hardly wider than the dimension of the passage exit perpendicular to the gas flow. Limitations on the number, size, and spacing of cooling passages results in gaps in the protective film and/or areas of low film cooling effectiveness which may produce localized hot spots. Airfoil hot spots are one factor which limits the operating temperature of the engine.
U.S. Pat. No. 3,527,543 to Howald uses divergently tapered passages of circular cross section to increase the entrainment of coolant in the boundary layer from a given passage. The passages are also preferably oriented in a plane extending in the longitudinal direction or partially toward the gas flow direction to spread the coolant longitudinally upon its exit from the passage as it moves downstream. Despite these features, it has been determined by smoke flow visualization tests and engine hardware inspection that the longitudinal width of the coolant film from an eliptical passage breakout (i.e. Howald) continues to expand longitudinally only about a maximum of one passage exit minor diameter after the coolant is ejected on the airfoil surface. This fact, coupled with typical longitudinal spacing of three to six diameters between passages, result in areas of airfoil surface between and downstream of longitudinally spaced passages which receive no cooling fluid from that row of passages. Conical, angled passages as described in Howald U.S. Pat. No. 3,527,543 provide at best probably no more than 70% coverage (percentage of the distance between the centers of adjacent hole breakouts which is covered by coolant).
The velocity of the air leaving the cooling passage is dependent on the ratio of its pressure at the passage inlet to the pressure of the gas stream at the passage outlet. In general the higher the pressure ratio, the higher the exit velocity. Too high an exit velocity results in the cooling air penetrating into the gas stream and being carried away without providing effective film cooling. Too low a pressure ratio will result in gas stream ingestion into the cooling passage causing a complete loss of local airfoil cooling. Total loss of airfoil cooling usually has disastrous results, and because of this a margin of safety is usually maintained. This extra pressure for the safety margin drives the design toward the high pressure ratios. Tolerance of high pressure ratios is a desirable feature of film cooling designs. Diffusion of the cooling air flow by tapering the passage, as in the Howald patent discussed above is beneficial in providing this tolerance, but the narrow diffusion angles taught therein (12.degree. maximum included angle) require long passages and, therefore, thick airfoil walls to obtain the reductions in exit velocities often deemed most desirable to reduce the sensitivity of the film cooling design to pressure ratio. The same limitation exists with respect to the trapezoidally shaped diffusion passages described in Sidenstick, U.S. Pat. No. 4,197,443. The maximum included diffusion angles taught therein in two mutually perpendicular planes are 7.degree. and 14.degree., respectively, in order to assure that separation of the cooling fluid from the tapered walls does not occur and the cooling fluid entirely fills the passage as it exits into the hot gas stream. With such limits on the diffusing angles, only thicker airfoil walls and angling of the passages in the airfoil spanwise direction can produce wider passage outlets and smaller gaps between passages in the longitudinal direction. Wide diffusion angles would be preferred instead, but cannot be achieved using prior art teachings.
Japanese Pat. No. 55-114806 shows, in its FIGS. 2 and 3 (reproduced herein as prior art FIGS. 15 and 16), a hollow airfoil having straight cylindrical passages disposed in a longitudinal row and emptying into a longitudinally extending slot formed in the external surface of the airfoil. While that patent appears to teach that the flow of cooling fluid from adjacent passages blends to form a film of cooling fluid of uniform thickness over the full length of the slot by the time the cooling fluid exits the slot and reaches the airfoil surface, our test experience indicates that the coolant fluid from the cylindrical passages moves downstream as a stripe of essentially constant width, which is substantially the diameter of the passage. Any diffusion which results in blending of adjacent stripes of coolant fluid occurs so far downstream that film cooling effectiveness at that point is well below what is required for most airfoil designs.
U.S. Pat. No. 3,515,499 to Beer et al describes an airfoil made from a stack of etched wafers. The finished airfoil includes several areas having a plurality of longitudinally spaced apart passages leading from an internal cavity to a common, longitudinally extending slot from which the cooling air is said to issue to form a film of cooling air over the airfoil external surface. In FIG. 1 thereof each passage appears to converge from its inlet to a minimum cross-sectional area where it intersects the slot. In the alternate embodiment of FIG. 9, the passage appears to have a small, constant size which exits into a considerably wider slot. Both configurations are likely to have the same drawbacks as discussed with respect to the Japanese patent; that is, the cooling fluid will not uniformly fill the slot before it enters the main gas stream, and considerably less than 100% film coverage downstream of the slot is likely.
Other publications relating to film cooling the external surface of an airfoil are: U.S. Pat. Nos. 2,149,510; 2,220,420; 2,489,683; and "Flight and Aircraft Engineer", No. 2460, Vol. 69, 3/16/56, pp. 292-295, all of which show the use of longitudinally extending slots for cooling either the leading edge or pressure and suction side airfoil surfaces. The slots shown therein extend completely through the airfoil wall to communicate directly with an internal cavity. Such slots are undesireable from a structural strength viewpoint; and they also require exceedingly large flow rates.
U.S. Pat. No. 4,303,374 shows a configuration for cooling the exposed, cut-back surface of the trailing edge of an airfoil. The configuration includes a plurality of longitudinally spaced apart, diverging passages within the trailing edge. Adjacent passages meet at their outlet ends to form a continuous film of cooling air over the cut-back surface.
A serial publication, "Advances in Heat Transfer", edited by T. F. Irvine, Jr. and J. P. Hartnett, Vol. 7, Academic Press (N.Y. 1971) includes a monograph titled Film Cooling, by Richard J. Goldstein, at pp. 321-379, which presents a survey of the art of film cooling. The survey shows elongated slots of different shapes extending entirely through the wall being cooled, and also passages of circular cross section extending through the wall.